An Energy-Efficient Hardware Implementation of HOG-Based Object Detection at 1080HD 60 fps with Multi-Scale Support
A real-time and energy-efficient multi-scale object detector hardware implementation is presented in this paper. Detection is done using Histogram of Oriented Gradients (HOG) features and Support Vector Machine (SVM) classification. Multi-scale detection is essential for robust and practical applica...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Springer-Verlag
2015
|
| Online Access: | http://hdl.handle.net/1721.1/100250 https://orcid.org/0000-0003-4841-3990 https://orcid.org/0000-0002-0376-4220 |
| _version_ | 1826202550203842560 |
|---|---|
| author | Sze, Vivienne Suleiman, Amr AbdulZahir |
| author2 | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
| author_facet | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Sze, Vivienne Suleiman, Amr AbdulZahir |
| author_sort | Sze, Vivienne |
| collection | MIT |
| description | A real-time and energy-efficient multi-scale object detector hardware implementation is presented in this paper. Detection is done using Histogram of Oriented Gradients (HOG) features and Support Vector Machine (SVM) classification. Multi-scale detection is essential for robust and practical applications to detect objects of different sizes. Parallel detectors with balanced workload are used to increase the throughput, enabling voltage scaling and energy consumption reduction. Image pre-processing is also introduced to further reduce power and area costs of the image scales generation. This design can operate on high definition 1080HD video at 60 fps in real-time with a clock rate of 270 MHz, and consumes 45.3 mW (0.36 nJ/pixel) based on post-layout simulations. The ASIC has an area of 490 kgates and 0.538 Mbit on-chip memory in a 45 nm SOI CMOS process. |
| first_indexed | 2024-09-23T12:09:16Z |
| format | Article |
| id | mit-1721.1/100250 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T12:09:16Z |
| publishDate | 2015 |
| publisher | Springer-Verlag |
| record_format | dspace |
| spelling | mit-1721.1/1002502022-09-28T00:34:25Z An Energy-Efficient Hardware Implementation of HOG-Based Object Detection at 1080HD 60 fps with Multi-Scale Support Sze, Vivienne Suleiman, Amr AbdulZahir Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Sze, Vivienne Sze, Vivienne Suleiman, Amr AbdulZahir A real-time and energy-efficient multi-scale object detector hardware implementation is presented in this paper. Detection is done using Histogram of Oriented Gradients (HOG) features and Support Vector Machine (SVM) classification. Multi-scale detection is essential for robust and practical applications to detect objects of different sizes. Parallel detectors with balanced workload are used to increase the throughput, enabling voltage scaling and energy consumption reduction. Image pre-processing is also introduced to further reduce power and area costs of the image scales generation. This design can operate on high definition 1080HD video at 60 fps in real-time with a clock rate of 270 MHz, and consumes 45.3 mW (0.36 nJ/pixel) based on post-layout simulations. The ASIC has an area of 490 kgates and 0.538 Mbit on-chip memory in a 45 nm SOI CMOS process. Texas Instruments Incorporated United States. Defense Advanced Research Projects Agency (Young Faculty Award Grant N66001-14-1-4039) 2015-12-14T22:36:25Z 2015-12-14T22:36:25Z 2015-12 2015-07 Article http://purl.org/eprint/type/JournalArticle 1939-8018 1939-8115 http://hdl.handle.net/1721.1/100250 Suleiman, Amr, and Vivienne Sze. "An Energy-Efficient Hardware Implementation of HOG-Based Object Detection at 1080HD 60 fps with Multi-Scale Support." Journal of Signal Processing Systems (December 2015). https://orcid.org/0000-0003-4841-3990 https://orcid.org/0000-0002-0376-4220 en_US http://dx.doi.org/10.1007/s11265-015-1080-7 Journal of Signal Processing Systems Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Springer-Verlag Sze |
| spellingShingle | Sze, Vivienne Suleiman, Amr AbdulZahir An Energy-Efficient Hardware Implementation of HOG-Based Object Detection at 1080HD 60 fps with Multi-Scale Support |
| title | An Energy-Efficient Hardware Implementation of HOG-Based Object Detection at 1080HD 60 fps with Multi-Scale Support |
| title_full | An Energy-Efficient Hardware Implementation of HOG-Based Object Detection at 1080HD 60 fps with Multi-Scale Support |
| title_fullStr | An Energy-Efficient Hardware Implementation of HOG-Based Object Detection at 1080HD 60 fps with Multi-Scale Support |
| title_full_unstemmed | An Energy-Efficient Hardware Implementation of HOG-Based Object Detection at 1080HD 60 fps with Multi-Scale Support |
| title_short | An Energy-Efficient Hardware Implementation of HOG-Based Object Detection at 1080HD 60 fps with Multi-Scale Support |
| title_sort | energy efficient hardware implementation of hog based object detection at 1080hd 60 fps with multi scale support |
| url | http://hdl.handle.net/1721.1/100250 https://orcid.org/0000-0003-4841-3990 https://orcid.org/0000-0002-0376-4220 |
| work_keys_str_mv | AT szevivienne anenergyefficienthardwareimplementationofhogbasedobjectdetectionat1080hd60fpswithmultiscalesupport AT suleimanamrabdulzahir anenergyefficienthardwareimplementationofhogbasedobjectdetectionat1080hd60fpswithmultiscalesupport AT szevivienne energyefficienthardwareimplementationofhogbasedobjectdetectionat1080hd60fpswithmultiscalesupport AT suleimanamrabdulzahir energyefficienthardwareimplementationofhogbasedobjectdetectionat1080hd60fpswithmultiscalesupport |